This paper applies the multi-agent system paradigm to collaborative negotiation in supply chain network. Multi-agent computational environments are suitable for dealing with a class of coordination and negotiation issues involving multiple autonomous or semiautonomous problem solving agents. An evolution teamwork system based on multi-agents that can organize most team members in supply chain network was proposed. The proposed model performs adaptive development relying on differential evolution process. The experimental results show that our developing teamwork system is able to provide the adaptability of team differential evolution is global optimization and continuously develop teamwork members for the resources management in supply chain network.